Explore the vast range of titles available.

Tweek signals are a special type of VLF (very low frequency) pulse, originally produced by lightning discharge, which becomes dispersive after repetitive bounces within the waveguide between the Earth’s surface and lower ionosphere. As such, tweek signals carry critical information about the region near the reflection height of the VLF waves, namely the D-region ionosphere. Although tweek measurements have been widely utilized in studies of the D-region ionosphere and lightning discharge, few statistical studies have been conducted, mainly due to the difficulty of manually identifying tweek signals from the enormous amount of VLF data with heavy noise. Considering the importance of tweek signals and the lack of a high-precision detection model, in this study, we propose a method to automatically and accurately pick out tweek signals from VLF measurements. This method is explicitly developed based on the you only look once (YOLO) model and a post-tracing process. Using a total of 2495 randomly selected VLF spectrogram images as the testing set, we evaluated the performance of this method. The precision and recall are found to be 92.0% and 89.2% for the first-order mode, and 97.5% and 86.7% for the first-two-order mode tweek, respectively. The time needed to process 10-s VLF measurements with a cadence of 4 μs is only 6.5 s, allowing for identifying the tweek signals from continuous VLF measurements in real time. Therefore, this method represents a reliable means to automatically detect tweek signals and enables the opportunity to statistically investigate the D-region ionosphere and lightning discharge via these signals.

Earthquake prediction is still a large challenge worldwide so far. In this paper, an automatic detection method was put into service immediately after the Wushi MS 7.1 earthquake on 23 January 2024 to weekly detect possible CSES (China Seismo-Electromagnetic Satellite) precursory information before impending aftershocks. An electron perturbation with an enhanced magnitude of 38.3% was first detected on 24 January 2024 at night orbit 33175 and the corresponding variations in different plasma parameters measured at this orbit presented a typical feature of electron depletion or plasma bubble with an abrupt decrease and then an increase after one minute. The Kp index was also checked during this period and the values reached 3.7 once on 23 and 24 January, which indicates that these ionospheric variations probably originated from solar activities instead of three strong aftershocks with a magnitude more than five in the following three days. However, uncertainties still exist. Then, an electron perturbation with amplitude of 24.6%, as well as an O+ one of 27.3%, was successfully searched automatically at the same revisiting orbit 33251 on 3 February 2024 in a magnetically quiet period. These two plasma variations, as well as ones of other ionospheric parameters, were characterized by highly synchronous properties, which increase the availability as seismic precursors. However, no obvious variations were observed at other revisiting orbits or other orbits near the aftershock areas during this period. An aftershock with magnitude of MS 5.3 and the strongest one of MS 5.8 took place on 24 and 25 February, respectively, 20 days after and 1000 km away.

In this investigation, 117718 ionospheric perturbations with a space size t = 20–300 s, but without a limit on amplitude (A), were automatically searched globally via software using ion density data measured by the Detection of Electromagnetic Emissions Transmitted from Earthquake Regions (DEMETER) satellite for approximately six years. After eliminating 18169 perturbations (PERs) that occurred in perturbed time (Kp ≥ 3) in order to avoid the global effects of geomagnetic storms, 99549 PERs were left. These PERs were divided into six groups with respect to local time and season, but only the summer and winter PERs were kept for good comparison. Each group of PERs was distributed on a map in terms of amplitude and space size. The distributions of all groups of plasma PERs showed that the equatorial ionization anomaly (EIA) structure developed well during the winter daytime. This structure is characterized by a low plasma PER density (Np) on both sides of the magnetic equator at low latitudes and a sudden enhancement of Np at a magnetic latitude of 15° at both sides of the magnetic latitudes. At the same time, this daytime anomaly was also clearly presented with a simultaneous enhancement in both Np and space size, even beyond 200 s along its boundary line. From the viewpoint of space size, there is a demarcation point, t = 120 s; the occurrence probabilities of day PERs are always higher than those at night before this point, whereas this result is reversed after this point. Another key point is that large spatial-scale perturbations (t > 200 s) present their major significance during daytime, especially during summer daytime, probably due to ionization under strong sunlight. The WSA (Wedell Sea Anomaly) phenomenon shows a clear configuration at all times except on the summer day. The topside ionosphere remained calm during the day in summer, and no positive PERs with a magnitude of more than 50% or a negative magnitude of more than 100% occurred in this WSA area. In winter nighttime, this WSA structure was characterized by major large positive PERs with a magnitude of more than 100%.

In this study, 117,718 ionospheric perturbations, with a space size (t) of 20–300 s but no amplitude (A) limit, were automatically globally searched via software utilizing ion density data measured by the DEMETER satellite for over 6 years. The influence of geomagnetic storms on the ionosphere was first examined. The results demonstrated that storms can globally enhance positive ionospheric irregularities but rarely induce plasma variations of more than 100%. The probability of PERs with a space size falling in 200–300 s (1400–2100 km if a satellite velocity of 7 km/s is considered) occurring in a geomagnetically perturbed period shows more significance than that in a quiet period. Second, statistical work was performed on ion PERs to check their dependence on local time, and it was shown that 24.8% of the perturbations appeared during the daytime (10:30 LT) and 75.2% appeared during the nighttime (22:30 LT). Ionospheric fluctuations with an absolute amplitude of A < 10% tend to be background variations, and the percentages of positive perturbations with a small A < 20% occur at an amount of 64% during the daytime and 26.8% during the nighttime, but this number is reversed for mid–large-amplitude PERs. Large positive PERs with A > 100% mostly occurred at night and negative ones with A < −100% occurred entirely at night. There was a demarcation point in the space size of t = 120 s, and the occurrence probabilities of day PERs were always higher than that of nighttime ones before this point, while this trend was contrary after this point. Finally, distributions of PERs according to different ranges of amplitude and space scale were characterized by typical seasonal variations either in the daytime or nighttime. EIA only exists in the dayside equinox and winter, occupying two low-latitude crests with a lower Np in both hemispheres. Large WSAs appear within all periods, except for dayside summer, and are full of PERs with an enhanced amplitude, especially on winter nights. The WN-like structure is obvious during all seasons, showing large-scale space. On the other hand, several magnetically anomalous zones of planetary-scale non-dipole fields, such as the SAMA, Northern Africa anomaly, and so on, were also successfully detected by extreme negative ion perturbations during this time.

Given the large volume of remote sensing images collected daily, automatic object detection and segmentation have been a consistent need in Earth observation (EO). However, objects of interest vary in shape, size, appearance, and reflecting properties. This is not only reflected by the fact that these objects exhibit differences due to their geographical diversity but also by the fact that these objects appear differently in images collected from different sensors (optical and radar) and platforms (satellite, aerial, and unmanned aerial vehicles (UAV)). Although there exists a plethora of object detection methods in the area of remote sensing, given the very fast development of prevalent deep learning methods, there is still a lack of recent updates for object detection methods. In this paper, we aim to provide an update that informs researchers about the recent development of object detection methods and their close sibling in the deep learning era, instance segmentation. The integration of these methods will cover approaches to data at different scales and modalities, such as optical, synthetic aperture radar (SAR) images, and digital surface models (DSM). Specific emphasis will be placed on approaches addressing data and label limitations in this deep learning era. Further, we survey examples of remote sensing applications that benefited from automatic object detection and discuss future trends of the automatic object detection in EO.

In the era of the first Industrial Revolution, many buildings were built with red bricks, and the heritage buildings built at that time are more than 100 years old. In these old heritage buildings, damage is bound to occur due to chemical and physical effects. Technologies such as automatic damage detection can effectively manage damage, but they can be affected by other categories present in heritage buildings. Therefore, this paper proposes a CNN algorithm that can automatically detect cracks and damage that occur in heritage buildings, as well as multi-label classification, such as doors, windows, arches, artwork, brick walls, stonewalls, and vents. A total of 2400 thermal infrared images are collected for 8 categories and automatic classification was performed using the CNN algorithm. The average precision and average sensitivity for the eight categories of heritage buildings are 97.72% and 97.43%, respectively. This paper defines the causes of misclassification as the following two causes: misclassification by multiple objects and misclassification by the perception of the CNN algorithm.

Automatic ship detection in SAR images plays an essential role in both military and civilian fields. However, most of the existing deep learning detection methods introduce complex models and huge calculations while improving the detection accuracy, which is not conducive to the application of real-time ship detection. To solve this problem, an efficient lightweight network YOLO-Lite is proposed for SAR ship detection in this paper. First, a lightweight feature enhancement backbone (LFEBNet) is designed to reduce the amount of calculation. Additionally, a channel and position enhancement attention (CPEA) module is constructed and embedded into the backbone network to more accurately locate the target location by capturing the positional information. Second, an enhanced spatial pyramid pooling (EnSPP) module is customized to enhance the expression ability of features and address the position information loss of small SAR ships in high-level features. Third, we construct an effective multi-scale feature fusion network (MFFNet) with two feature fusion channels to obtain feature maps with more position and semantic information. Furthermore, a novel confidence loss function is proposed to effectively improve the SAR ship target detection accuracy. Extensive experiments on SSDD and SAR ship datasets verify the effectiveness of our YOLO-Lite, which can not only accurately detect SAR ships in different backgrounds but can also realize a lightweight architecture with low computation cost.

Observational studies of human behaviour often require the annotation of objects in video recordings. Automatic object detection has been facilitated strongly by the development of YOLO (‘you only look once’) and particularly by YOLOv8 from Ultralytics, which is easy to use. The present study examines the conditions required for accurate object detection with YOLOv8. The results show almost perfect object detection even when the model was trained on a small dataset (100 to 350 images). The detector, however, does not extrapolate well to the same object in other backgrounds. By training the detector on images from a variety of backgrounds, excellent object detection can be restored. YOLOv8 could be a game changer for behavioural research that requires object annotation in video recordings.

Storms can cause significant damage to forest areas, affecting biodiversity and infrastructure and leading to economic loss. Thus, rapid detection and mapping of windthrows are crucially important for forest management. Recent advances in computer vision have led to highly-accurate image classification algorithms such as Convolutional Neural Network (CNN) architectures. In this study, we tested and implemented an algorithm based on CNNs in an ArcGIS environment for automatic detection and mapping of damaged areas. The algorithm was trained and tested on a forest area in Bavaria, Germany. . It is a based on a modified U-Net architecture that was optimized for the pixelwise classification of multispectral aerial remote sensing data. The neural network was trained on labeled damaged areas from after-storm aerial orthophotos of a ca. 109 k m 2 forest area with RGB and NIR bands and 0.2-m spatial resolution. Around 10 7 pixels of labeled data were used in the process. Once the network is trained, predictions on further datasets can be computed within seconds, depending on the size of the input raster and the computational power used. The overall accuracy on our test dataset was 92 % . During visual validation, labeling errors were found in the reference data that somewhat biased the results because the algorithm in some instance performed better than the human labeling procedure, while missing areas affected by shadows. Our results are very good in terms of precision, and the methods introduced in this paper have several additional advantages compared to traditional methods: CNNs automatically detect high- and low-level features in the data, leading to high classification accuracies, while only one after-storm image is needed in comparison to two images for approaches based on change detection. Furthermore, flight parameters do not affect the results in the same way as for approaches that require DSMs and DTMs as the classification is only based on the image data themselves, and errors occurring in the computation of DSMs and DTMs do not affect the results with respect to the z component. The integration into the ArcGIS Platform allows a streamlined workflow for forest management, as the results can be accessed by mobile devices in the field to allow for high-accuracy ground-truthing and additional mapping that can be synchronized back into the database. Our results and the provided automatic workflow highlight the potential of deep learning on high-resolution imagery and GIS for fast and efficient post-disaster damage assessment as a first step of disaster management.

Surveying threatened and invasive species to obtain accurate population estimates is an important but challenging task that requires a considerable investment in time and resources. Estimates using existing ground-based monitoring techniques, such as camera traps and surveys performed on foot, are known to be resource intensive, potentially inaccurate and imprecise, and difficult to validate. Recent developments in unmanned aerial vehicles (UAV), artificial intelligence and miniaturized thermal imaging systems represent a new opportunity for wildlife experts to inexpensively survey relatively large areas. The system presented in this paper includes thermal image acquisition as well as a video processing pipeline to perform object detection, classification and tracking of wildlife in forest or open areas. The system is tested on thermal video data from ground based and test flight footage, and is found to be able to detect all the target wildlife located in the surveyed area. The system is flexible in that the user can readily define the types of objects to classify and the object characteristics that should be considered during classification.